CN112513700B

CN112513700B - Optical fiber circuit, module, and optical fiber circuit manufacturing method

Info

Publication number: CN112513700B
Application number: CN201980050173.5A
Authority: CN
Inventors: 铃木雅人; 山本义典; 田村欣章; 长谷川健美
Original assignee: Sumitomo Electric Industries Ltd
Current assignee: Sumitomo Electric Industries Ltd
Priority date: 2018-08-24
Filing date: 2019-08-21
Publication date: 2022-10-28
Anticipated expiration: 2039-08-21
Also published as: WO2020040220A1; CN112513700A; US20210157053A1; US11493690B2

Abstract

The optical fiber line of one embodiment of the present invention includes HNLF, SMF, and MFD transitions. The MFD transition portion is formed by the end portions of HNLF and SMF butted at the weld point of HNLF and SMF, and is a section in which MFD changes so that the difference between the maximum value and the minimum value becomes 0.3 μm or more at a distance of 100 μm. The connection loss at 1550nm of HNLF and SMF is less than one fifth of the ideal butt loss at the constant portions of both HNLF and SMF. The MFD transition portion has a total length of 10mm or less. In a region between one end face located at the weld point and the other end face located at a distance of 50 μm or more and 300 μm or less from the one end face among the end portions of HNLF, the MFD monotonically increases from the other end face to the one end face.

Description

Optical fiber circuit, module, and optical fiber circuit manufacturing method

Technical Field

The present disclosure relates to an optical fiber circuit, a module, and an optical fiber circuit manufacturing method.

The present application claims priority from japanese patent application No. 2018-157023, filed 24/8/2018, which is incorporated into the present specification on the basis of its content and by reference to its entirety.

Background

Various configurations of optical fiber lines obtained by fusion-splicing the ends of different types of optical fibers are known.

The optical Fiber line described in patent document 1 is a line in which a single mode optical Fiber (single mode optical Fiber) having an increased effective area and a Dispersion Compensating optical Fiber (DCF) are fusion-connected. In the optical fiber line manufacturing method described in patent document 1, after fusion-splicing is performed by electric discharge, a fusion-spliced portion is heated while applying tension, thereby reducing a connection loss.

The optical Fiber line described in patent document 2 is a line in which a Single Mode optical Fiber (SMF) and a DCF are fusion-spliced. In the method for manufacturing an optical fiber circuit described in patent document 2, a thermal-diffused Expanded Core (TEC) is formed by heating a fusion-spliced portion with flame after fusion splicing, and thus a reduction in the splice loss is attempted.

The optical fiber lines described in patent documents 3 and 4 are those in which short bridge fibers are inserted between the SMF and the DCF and these fibers are fusion-spliced.

Documents of the prior art

Patent document

Patent document 1: japanese patent laid-open publication No. 2002-214467;

patent document 2: japanese patent laid-open publication No. 2006-506686;

patent document 3: japanese patent laid-open publication No. 8-190030;

patent document 4: japanese patent laid-open publication No. 2006-99147.

Non-patent literature

Non-patent document 1: marcuse, "Loss Analysis of Single-Mode Fiber spots", the Bell System Technical Journal, vol.56, no. 5, pp.703-718, 1977.

Disclosure of Invention

An optical fiber circuit according to an embodiment of the present disclosure includes: a Highly nonlinear Fiber (HNLF: high Non-Linear optical Fiber), a Single Mode Fiber (SMF), and an MFD transition section where the Mode Field Diameter (MFD: mode Field Diameter) of propagating light changes along the length direction of the Fiber. HNLF includes: a core, a recess, and a cladding. The recess portion surrounds the core portion and has a refractive index lower than that of the core portion. The cladding surrounds the recess and has a refractive index lower than that of the core and higher than that of the recess. The SMF is melt-connected with HNLF and includes a core and a cladding surrounding the core and having a refractive index lower than that of the core. The MFD transition portion is formed by the end portions of HNLF and SMF that are butted with each other with the weld point of HNLF and SMF interposed therebetween. The MFD transition portion is a section in which the MFD of the wavelength used changes so that the difference between the maximum value and the minimum value becomes 0.3 μm or more at a distance of 100 μm in the longitudinal direction of the optical fiber (the longitudinal direction of the optical fiber line). The connection loss at 1550nm wavelength of HNLF and SMF is less than one fifth of the ideal butt loss calculated based on the mismatch between the MFDs at 1550nm wavelength at the constant portions of both HNLF and SMF except for the MFD transition. The MFD transition portion has a predetermined total length along the longitudinal direction of the optical fiber of 10mm or less. Further, in a region between a first surface located at the weld point and a second surface located at a distance of 50 μm or more and 300 μm or less from the first surface in the end portion of the HNLF, the MFD of the wavelength used monotonically increases from the second surface to the first surface.

Drawings

Fig. 1 is a diagram showing an experimental system for measuring a connection loss of an optical fiber line before and after fusion connection between HNLF and SMF.

Fig. 2 is a diagram showing an example of a reinforcing structure of a weld of HNLF and SMF.

Fig. 3 is a graph showing a refractive index distribution of HNLF (constant portion).

Fig. 4 is a table summarizing parameters of SMF.

Fig. 5 is a table summarizing the parameters of HNLF.

Fig. 6 is a flowchart for explaining the optical fiber circuit manufacturing method.

Fig. 7 is a table summarizing the manufacturing conditions and connection loss of the optical fiber lines of the samples of the present disclosure, and the manufacturing conditions and connection loss of each of the optical fiber lines of comparative examples 1 to 3.

Fig. 8 is a graph showing a relationship between the total additional discharge time in the additional discharge step and the connection loss at each connection point for each optical fiber line of the sample of the present disclosure, comparative example 2, and comparative example 3.

Fig. 9 is a graph showing a change in MFD along the optical fiber length direction (the length direction of the optical fiber line) for each of the optical fiber lines of the sample of the present disclosure, comparative example 1, and comparative example 2.

FIG. 10 shows GeO as a typical dopant ₂ And F, and the temperature dependence of the diffusion rate.

FIG. 11 shows the diffusion velocity ratio (v) _F /v _G ) Graph of the relationship between heating time at 20 or 30 f and MFD of HNLF.

Fig. 12 is a graph for explaining a change in the shape of the refractive index profile of HNLF in the heated state shown by the curve G1110 and the curve G1120 in fig. 11.

Fig. 13 is a graph showing the results of evaluating the mechanical strength characteristics of each of the optical fiber lines of the sample of the present disclosure and comparative example 2 as a weibull distribution curve.

Detailed Description

[ problem to be solved by the present disclosure ]

DCF is preferably used in terms of effective area (A) for suppressing nonlinearity _eff : effective area) is large. On the other hand, HNLF (high nonlinear fiber) is preferably a for efficiently exciting the nonlinear effect _eff Is small. Therefore, the MFD (mode field diameter) of HNLF is generally smaller than that of DCF. The MFD difference between HNLF and SMF is greater than the MFD difference between DCF and SMF. Therefore, it is difficult to perform low-loss fusion bonding of HNLF and SMF.

In the case where dopant diffusion is performed by flame heating after fusion-splicing in order to reduce the connection loss between fusion-spliced optical fibers, it is difficult to improve productivity. This is because the maximum reach temperature of the flame is low. That is, this is because the time required to increase the MFD of the optical fiber by diffusing the dopant contained in the optical fiber and the time required to match the MFD between fusion-spliced optical fibers are long.

On the other hand, the maximum reaching temperature can be increased by the doping diffusion by the discharge heating as compared with the doping diffusion by the flame heating, and therefore, the processing time is short and the productivity is high. However, in order to reduce the fusion loss, it is necessary to make the MFD change into a tapered shape at the MFD transition portion located in the optical fiber connection region (a state in which the MFD changes monotonically along the optical fiber length direction, hereinafter referred to as "MFD taper"). The MFD cone is less likely to be formed by the doping diffusion by the discharge heating than by the flame heating, and it is difficult to reduce the connection loss.

When the amount of F per unit length (cross-sectional area × concentration of the portion to which F is added) (referred to as "fluorine (F) area concentration" in the present specification) is small, the effect of increasing the MFD by thermal diffusion of F is limited. Since the diffusion rate of Ge contained in the core portion is slower than that of F, it takes time to reduce the fusion loss. The core and the cladding do not substantially contain a dopant (refractive index lowering material) for lowering the refractive index, and if the amount of F contained in the recess is limited, it takes time to reduce the connection loss.

When a short bridge fiber is inserted between optical fibers to be connected, although it is effective in reducing connection loss, productivity is reduced because one fusion-spliced portion is increased.

The present disclosure provides an optical fiber circuit having a structure in which HNLF and SMF are fusion-connected to each other, having a small connection loss, and being capable of being manufactured with high productivity. Additionally, the present disclosure provides a method of enabling the manufacture of such an optical fiber circuit.

[ Effect of the present disclosure ]

The optical fiber circuit of the present disclosure has a structure in which HNLF and SMF are fusion-connected to each other, has a small connection loss, and can be manufactured with high productivity.

[ description of embodiments of the present disclosure ]

First, the contents of the embodiments of the present invention will be described separately.

(1) One aspect of the fiber circuit of one embodiment of the present disclosure has HNLF (high nonlinear fiber), SMF (single mode fiber), and MFD transition. HNLF includes: a core, a recess, and a cladding. The recess portion surrounds the core portion and has a refractive index lower than that of the core portion. The cladding surrounds the recess and has a refractive index lower than that of the core and higher than that of the recess. The SMF is melt-connected with HNLF and includes a core and a cladding surrounding the core and having a refractive index lower than that of the core. The MFD transition portion is formed by the ends of HNLF and SMF that butt at the weld point of HNLF and SMF. The MFD transition is a section in which the MFD of the wavelength used changes so that the difference between the maximum value and the minimum value is 0.3 μm or more over a distance of 100 μm along the longitudinal direction of the optical fiber (the longitudinal direction of the optical fiber line). The connection loss of HNLF and SMF at a wavelength of 1550nm is less than one fifth of the ideal butt loss calculated based on the mismatch between the MFDs of the constant portions other than the MFD transition portions of both HNLF and SMF at the wavelength of 1550 nm. The MFD transition portion has a predetermined total length along the longitudinal direction of the optical fiber of 10mm or less. In addition, in the region between the first surface located at the welding point and the second surface located at a distance of 50 μm or more and 300 μm or less from the first surface in the end portion of HNLF, the MFD of the wavelength used monotonically increases from the second surface to the first surface.

Further, in the present disclosure, "HNLF (high nonlinear fiber)" means that the nonlinear coefficient at a wavelength of 1550nm is 7W ^-1 ·km ^-1 Above and A _eff (effective area) 30 μm ² The following optical fibers. The "constant part" is a section in which the difference between the maximum value and the minimum value of MFD of the wavelength used at a distance of 100 μm along the length direction of the optical fiber converges in a range of less than 0.3 μm; the "MFD transition" is a section in which the MFD of a wavelength used at a distance of 100 μm changes so that the difference between the maximum value and the minimum value becomes 0.3 μm or more. The "concentration" of the dopant is expressed by a mass ratio (a ratio of the mass of the object to the mass of the whole). In addition, "idealButt loss α _ideal [dB]"is a loss caused only by a mismatch of MFD at 1550nm wavelength of each of two optical fibers sandwiching a fusion-spliced point, and a value W obtained by dividing MFD at 1550nm wavelength of a reference waveguide mode of a constant portion of one optical fiber by MFD at 1550nm wavelength of a reference waveguide mode of a constant portion of the other optical fiber is used ₁₂ Expressed by the following formula (1):

α _ideal ＝-10LOg ₁₀ {4W ₁₂ ² /(W ₁₂ ² +1) ² } (1)

(see non-patent document 1).

(2) As one aspect of the present disclosure, HNLF is composed of quartz glass. In the constant portion of HNLF, the core portion does not substantially contain a refractive index lowering agent as a dopant for lowering the refractive index of the glass at a wavelength of 1550nm, and contains a refractive index raising agent as a dopant for raising the refractive index of the glass at a wavelength of 1550 nm. The recessed portion contains a refractive index lowering agent and does not substantially contain a refractive index raising agent. The cladding layer contains substantially no refractive index lowering agent. In addition, even when only the recessed portion substantially contains a dopant for lowering the refractive index, HNLF and SMF can be efficiently melt-connected with a low loss. Here, "substantially no dopant" means that the mass ratio of the dopant is less than 100ppm. Within this range, the various dopants do not produce a change in refractive index that can affect the optical characteristics of the optical fiber.

(3) As one aspect of the present disclosure, in the constant portion of HNLF, the core preferably contains GeO as a refractive index increasing agent ₂ . The depressed portion preferably contains a refractive index lowering agent having an area concentration of 0.4X 10 ⁶ ppm·μm ² Above and 3.2X 10 ⁶ ppm·μm ² The following are fluorine. In this case, even when the area concentration of F, which is a dopant for lowering the refractive index, contained in the depressed portion is a finite amount, HNLF and SMF can be fusion-connected with high efficiency and low loss.

(4) As one aspect of the present disclosure, it is preferable that in an end portion of HNLF, the outer diameter of the cladding of HNLF monotonically decreases or increases from a constant portion of the HNLF to a constant portion of SMF. In this case, there is no recess in the surface of the cladding, and therefore mechanical strength can be ensured.

(5) As one aspect of the present disclosure, the overall length of the MFD transition portion may be 2mm or less. The MFD transition portion may have a predetermined length in the fiber longitudinal direction of a portion constituting HNLF of 1.5mm or less. In this case, since the amount of heat required to form the MFD transition portion can be reduced, low-loss connection can be efficiently performed.

(6) As one aspect of the present disclosure, the SMF is preferably an optical fiber conforming to g.657.A1 standard of ITU-T. Since the optical fiber conforming to the g.657.a1 standard is more resistant to bending than the general-purpose SMF, it can be housed in a module with a smaller radius (the module volume can be reduced).

(7) As one aspect of the present disclosure, it is preferable that in the constant portion of HNLF, the outer diameter of the core portion is 2.5 μm or more and 4.0 μm or less. Preferably, the ratio of the outer diameter of the recess to the outer diameter of the core is 1.8 or more and 3.2 or less. The relative refractive index difference between the maximum refractive index of the core and the average refractive index of the cladding is preferably 1.3% or more and 2.0% or less. The relative refractive index difference between the minimum refractive index of the depressed portion and the average refractive index of the clad is preferably-1.0% or more and-0.5% or less. Further, the constant portion of HNLF preferably includes: at 1550nm wavelength of 10 μm ² Above and 30 μm ² The following A _eff MFD at 1550nm having a wavelength of 3.5 μm or more and 5.0 μm or less, and-3.0 ps · km at 1550nm ^-1 ·nm ^-1 Above and 1.5ps km ^-1 ·nm ^-1 The following Chromatic dispersion (Chromatic dispersion) and a wavelength of 7W at 1550nm ^-1 ·km ^-1 Above and 20W ^-1 ·km ^-1 The following nonlinear coefficients. In this case, since the wavelength dispersion in the C band has flat optical characteristics, phase matching of a nonlinear phenomenon can be obtained in a wide frequency band in the C band.

(8) As one aspect of the present disclosure, it is preferable that the diffusion velocity v of fluorine in a region (a region between a first surface and a second surface which is distant from the first surface by 50 μm) included in the end portion of HNLF _F With GeO ₂ Velocity v of diffusion of _G Ratio of (v) _F /v _G ) Is 22 or more and 40 or less. Deformation (dishing) of the MFD transition portion can be avoided by not melting the fusion portion including the fusion point of HNLF and SMF.

(9) As one aspect of the present disclosure, it is preferable that a butt joint region, which is constituted by an end of the HNLF and an end of the SMF and includes a weld point, be protected by being covered with a resin or being housed in a reinforcing sleeve. Here, the butt joint region protected by the resin or the reinforcing sleeve preferably has a strength of 200kpsi (= 1.4 MPa) or more. This structure is effective for preventing breakage at high power light-on (e.g., 1W or more).

(10) As one aspect of the present disclosure, both or one of HNLF and SMF may have a covering resin layer surrounding the cladding layer and containing a colorant. In particular, in the structure in which both HNLF and SMF have the above-described covering resin layers, the covering resin layers of HNLF and SMF may also contain colorants different in color from each other, respectively. In this case, it is possible to prevent erroneous connection between SMFs and welding between HNLFs. In addition, the appearance inspection when the optical fibers are housed in the module is also easy.

(11) As one aspect of the present disclosure, the 50% breaking strength in the tensile strength test is preferably 4.5N or more.

(12) A module according to an embodiment of the present disclosure includes an optical fiber line (an optical fiber line according to an embodiment of the present disclosure) having the above-described configuration, and a metal case that houses the optical fiber line. With this configuration, it is possible to prevent the optical fiber from being damaged by external impact, heat, or the like during actual use. In addition, heat generated in the optical fiber connection portion can be efficiently released to the outside.

(13) The method for manufacturing an optical fiber circuit according to an embodiment of the present disclosure includes: a configuration step, a welding step, and an additional discharge step. In the disposing step, an end of one of HNLF and SMF and an end of the other are disposed so as to face each other in a butt joint state. Here, HNLF includes: a core, a recess, and a cladding. The recess portion surrounds the core portion and has a refractive index lower than that of the core portion. The cladding surrounds the recess and has a refractive index lower than that of the core and higher than that of the recess. The welding step is performed after the arrangement step. In this welding step, the butted ends of HNLF and SMF are melted by heating. Thereby, the respective ends of HNLF and SMF are welded. The additional discharging step is performed after the welding step. In the additional discharge step, a region located on the HNLF side with respect to the welding point of HNLF and SMF is reheated by discharge. By this additional discharge step, an optical fiber line (an optical fiber line according to an embodiment of the present disclosure) having the above-described configuration can be obtained.

(14) In one aspect of the present disclosure, in the additional discharge step, it is preferable that the discharge power for diffusing fluorine contained in HNLF is continuously reheated for 50 seconds or more in a region located on the HNLF side with respect to the welding point so as not to melt the cladding layer of HNLF. By heating the HNLF at a temperature at which the cladding layer of the HNLF does not melt, an increase in welding loss due to deformation of the cladding layer can be effectively avoided.

(15) In one aspect of the present disclosure, in the additional discharge step, it is preferable that the end portion of the HNLF and the end portion of the SMF which are melt-connected are reheated by discharge while the heating portion is moved relative to the end portion of the HNLF and the end portion of the SMF which are melt-connected. According to this configuration, even if the dopant is diffused by the electric discharge heating in which the heating range is partially, the heating can be performed substantially in a wide range.

(16) As one aspect of the present disclosure, in the additional discharge step, it is preferable that the reheating by discharge is performed in a range (end of HNLF) in which a length along the longitudinal direction of the optical fiber is 1 time or more of an outer diameter of the clad. According to this configuration, the discharge can be performed in a wide range by increasing the distance between the discharge electrode rods, thickening the discharge electrode rods, or the like. This can shorten the time required for additional discharge.

In the above, the aspects listed in this column [ description of embodiments of the present disclosure ] can be applied to each of the remaining aspects, or can be applied to all combinations of these remaining aspects.

[ details of embodiments of the present disclosure ]

Hereinafter, specific configurations of an optical fiber circuit, a module, and an optical fiber circuit manufacturing method according to an embodiment of the present disclosure will be described in detail with reference to the drawings. The present invention is not limited to these examples, but is defined by the scope of the claims, and is intended to include meanings equivalent to the scope of the claims and all modifications within the scope. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted.

The optical fiber line according to the present embodiment has a structure in which a highly nonlinear fiber (HNLF) and an SMF are fusion-spliced to each other. Fig. 1 is a diagram showing an experimental system for measuring a connection loss of an optical fiber line before and after fusion connection between HNLF and SMF.

Configuration 1 of fig. 1 (before welding) shows an experimental system for measuring Reference Power. In the experimental system of structure 1, the light of 1550nm wavelength output from an LD (Laser Diode) light source 41 was incident on the incident end of an SMF (single mode fiber) 10. The light emitted from the exit end of the SMF10 after being guided by the SMF10 is received by the optical power meter 42, and the optical power (reference power) of the received light is detected. In addition, in the configuration 1 of fig. 1, the position indicated by the arrow C1 is the cut-off position of the SMF10 that is cut off after the reference power is measured.

Next, the SMF10 is divided into the SMF11 and the SMF12 by cutting the SMF10 at a position indicated by an arrow C1, a highly nonlinear optical fiber (HNLF) 20 is inserted between the SMF11 and the SMF12, one end of the HNLF20 and the end of the SMF11 are fusion-connected at a fusion-connection point 31, and the other end of the HNLF20 and the end of the SMF12 are fusion-connected at a fusion-connection point 32. After the cutting and welding steps, the structure 2 (after welding) of fig. 1 is obtained. In the experimental system of configuration 2, the light of 1550nm wavelength output from the LD light source 41 was incident on the incident end of the SMF 11. The light emitted from the exit end of the SMF12 after being guided sequentially in the order of the SMF11, HNLF20, and SMF12 is received by the optical power meter 42 (received optical power is detected). As shown in configuration 2, the HNLF module is configured by housing the optical fiber line including SMF11, HNLF20, and SMF12 in the metal case 50.

Fig. 2 is a diagram showing an example of a reinforcing structure of a weld of HNLF and SMF. Specifically, the drawings show examples of the reinforcing structure of the welding point 31 between the SMF11 and the HNLF20, and the welding point 32 between the HNLF20 and the SMF 12.

Configuration 0 in fig. 2 shows a state before reinforcement (a state before the MFD transition portion is formed), specifically, a state after one end of the SMF11 and one end of the HNLF20 are fusion-connected along the central axis AX, or a state after the other end of the HNLF20 and one end of the SMF12 are fusion-connected along the central axis AX. In addition, since the connection structure between SMF11 and HNLF20 and the connection structure between HNLF20 and SMF12 are substantially the same, only the connection relationship between HNLF20 and SMF11 will be referred to in the following description.

The SMF10 is an optical fiber conforming to g.657.A1 standard of ITU-T, and the SMF11 and the SMF12 divided from the SMF10 are also optical fibers conforming to g.657.A1 standard. The SMF11 (SMF 12) is made of quartz glass, and has a core 101 extending along the central axis AX, a clad 102 surrounding the core 101, and a coating resin layer 103 surrounding the clad 102. The refractive index of the cladding 102 is lower than the refractive index (maximum refractive index) of the core 101. A part of the covering resin layer 103 was removed at each end of the SMF11 and the SMF12 to be melt-connected to the HNLF 20.

On the other hand, the HNLF20 is made of quartz glass, and has a core 201 extending along the central axis AX, a recess 202 surrounding the core 201, a cladding 203 surrounding the recess 202, and a covering resin layer 204 surrounding the cladding 203. The refractive index of the recess 202 is lower than the refractive index (maximum refractive index) of the core 201. The refractive index of the cladding 203 is lower than that of the core 201 (maximum refractive index) and higher than that of the recess 202 (minimum refractive index). At both ends of the HNLF20, a part of the covering resin layer 204 is removed to be melt-connected to the SMF11 and the SMF12, respectively.

In the example of fig. 2, the covering resin layers 204 and 103 are provided on both of the HNLF20 and the SMF11 (SMF 12), but only one of them may be provided. The covering resin layer 204 and the covering resin layer 103 each preferably contain a colorant, and in a structure in which both the HNLF20 and the SMF11 (SMF 12) are provided with the covering resin layers 204, 103, the covering resin layer 204 and the covering resin layer 103 preferably contain colorants different in color from each other, respectively.

The region SP shown in the structure 0 in fig. 2 indicates a constant portion of the SMF11 (SMF 12) and a constant portion of the HNLF 20. In particular, the constant portion SP of the HNLF20 has a W-type refractive index distribution as shown in fig. 3. On the other hand, an area TP shown by configuration 0 in fig. 2 indicates an area to be an MFD transition portion (MFD taper forming area). Substantially, the region TP coincides with a region indicating the MFD transition portion where the MFD cone is formed.

The region on the HNLF20 side (the region included in the region TP) is reheated by electric discharge at the welding point 31 (32) of the HNLF20 and the SMF11 (SMF 12), thereby forming an MFD transition portion in the region TP. Further, the reheating by the discharge is preferably performed at a discharge power for diffusing F (fluorine) contained in the HNLF20 without melting the cladding 203 of the HNLF20 for 50 seconds or more. Further, it is preferable to perform reheating by discharge while relatively moving the heating portion with respect to the end portions of the HNLF20 and the SMF11 (SMF 12). The length of the reheated region in the longitudinal direction of the optical fiber is preferably 1 time or more of the outer diameter of the cladding 203.

As described above, the MFD transition portion formed in the region TP is constituted by the end portions of both HNLF20 and SMF11 (SMF 12) abutting against each other with the fusion point 31 (32) of HNLF20 and SMF11 (SMF 12) therebetween. In the region between the first surface (the contact surface between HNLF20 and SMF11 (SMF 12)) located at the welding point 31 (32) and the second surface located at a distance of 50 μm or more and 300 μm or less from the first surface, MFD of the wavelength used monotonically increases from the second surface to the first surface (MFD taper) at the end of HNLF 20.

The MFD transition is defined as a section in which the MFD of the wavelength used at a distance of 100 μm along the longitudinal direction of the optical fiber (corresponding to the central axis AX) changes so that the difference between the maximum value and the minimum value becomes 0.3 μm or more. The total length of the MFD transition portion in the longitudinal direction of the optical fiber is 10mm or less, preferably 2mm or less. In a configuration in which the total length of the MFD transition portion is 2mm or less, the predetermined length of the portion of the MFD transition portion constituting a part of the HNLF20 in the optical fiber longitudinal direction is preferably 1.5mm or less. Further, by the MFD transition portion formed as described above, the connection loss at the wavelength of 1550nm of HNLF20 and SMF11 (SMF 12) becomes less than one fifth of the ideal butt loss calculated based on the mismatch between the MFDs at the wavelength of 1550nm in the constant portion SP of both HNLF20 and SMF11 (SMF 12). In the structure 0 in fig. 2, the outer diameter of the cladding 203 at the HNLF20 is fixed, but the outer diameter of the cladding 203 at the end may be monotonically decreased or increased from the constant portion SP of the HNLF20 to the constant portion SP of the SMF11 (SMF 12).

In the structure 1 in fig. 2, an MFD transition portion is formed at a butt region that is constituted by an end of the HNLF20 and an end of the SMF11 (SMF 12) and includes the welding point 31 (32). The vicinity of the fusion-spliced point of the optical fiber line is protected by covering the butted area with recoating resin 300. On the other hand, in the structure 2 of fig. 2, the butt area is housed in the reinforcing sleeve 400. In any configuration, the butt areas protected by either the recoating resin 300 or the reinforcing sleeve 400 have a strength of 200kpsi or more.

Fig. 3 is a graph showing a refractive index distribution of HNLF20 (in particular, the constant portion SP). HNLF20 has a W-type refractive index profile. That is, the constant portion SP of the HNLF20 has a core 201 (outer diameter a), a recess 202 (outer diameter b) surrounding the core 201, a clad 203 (outer diameter c) surrounding the recess 202, and a covering resin layer 204 surrounding the clad 203. The recess 202 has a refractive index lower than that of the core 201. The cladding 203 has a refractive index lower than that of the core 201 and higher than that of the recess 202. Typically, the nonlinear coefficient γ of HNLF20 at 1550nm wavelength is 7W ^-1 ·km ^-1 Above, A at a wavelength of 1550nm _eff Is 30 μm ² The following.

The SMF10 (substantially SMFs 11 and 12) and the HNLF20 are made of quartz glass. In either of HNLF20 and SMF11 (SMF 12), the constant portion SP is defined as a section in which the change in MFD of the wavelength used is limited to a difference between the maximum value and the minimum value of less than 0.3 μm over a distance of 100 μm along the longitudinal direction of the fiber. Particularly, in the constant portion SP of the HNLF20, the core portion 201 does not substantially contain a dopant (refractive index lowering agent) for lowering the refractive index of the glass at a wavelength of 1550nm, and contains, for example, geO as a dopant (refractive index raising agent) for raising the refractive index of the glass at a wavelength of 1550nm ₂ . On the other hand, the recess 202 is substantially absentContaining a refractive index increasing agent and, as a refractive index decreasing agent, for example, in an area concentration of 0.4X 10 ⁶ ppm·μm ² Above and 3.2X 10 ⁶ ppm·μm ² The following are fluorine. The cladding 203 contains substantially no refractive index lowering agent.

Preferred ranges for various characteristics of HNLF20 are as follows. The outer diameter a of the core 201 is 2 times the distance from the center (central axis AX) of the HNLF20 to the first position where the differential value of the refractive index distribution first becomes the minimum value. The outer diameter b of the recessed portion 202 is 2 times the distance from the center of HNLF20 to the second position where the differential value of the refractive index distribution becomes maximum in the region where the distance from the center of HNLF20 is greater than a/2. Specifically, the outer diameter a of the core 201 is 2.5 μm or more and 4.0 μm or less. The ratio of the outer diameters (b/a) of the recess 202 and the core 201 is 1.8 or more and 3.2 or less. Relative refractive index difference Δ between the maximum refractive index of the core 201 and the average refractive index of the cladding 203 _core Is 1.3% or more and 2.0% or less. Relative refractive index difference Δ between minimum refractive index of the recess 202 and average refractive index of the cladding 203 _dep Is more than-1.0% and less than-0.5%. A at 1550nm wavelength _eff Is 10 μm ² Above and 30 μm ² The following. MFD at a wavelength of 1550nm is 3.5 μm or more and 5.0 μm or less. The chromatic dispersion (chromatic dispersion) at a wavelength of 1550nm is-3.0 ps-km ^-1 ·nm ^-1 Above and 1.5ps km ^-1 ·nm ^-1 The following. The nonlinear coefficient gamma at 1550nm wavelength is 7W ^-1 ·km ^-1 Above and 20W ^-1 ·km ^-1 The following.

Fig. 4 is a table summarizing parameters of SMF10 (substantially SMF11 and SMF 12) used in the experiment described later. SMF10 is an optical fiber conforming to the ITU-T G.657.A1 standard. Fig. 5 is a table summarizing the parameters of HNLF20 used in the experiment described later.

Fig. 6 is a flow chart of a method of manufacturing an optical fiber circuit. The optical fiber line manufacturing method shown in fig. 6 is a method of connecting an end of an HNLF corresponding to the HNLF20 and an end of an SMF corresponding to the SMF11 and the SMF12 of the present disclosure, and the preparation step S1, the arrangement step S2, the alignment step S3, the fusion step S4, the additional discharge step S5, and the reinforcement step S6 are sequentially performed. The fusion-

splicing points

31 and 32 shown in the structure 2 (after fusion splicing) of fig. 1 are also connected by this optical fiber line manufacturing method.

In the preparation step S1, after removing a part of each of the coating resin layers in a predetermined range including the prepared end portions of the SMF and HNLF, the end portions of the SMF and HNLF are cut with a fiber cutter. The cutting angle of the end face is preferably 1.0 ° or less, more preferably 0.5 ° or less.

In the arranging step S2, the respective end portions of the SMF and the HNLF are arranged in the arc discharge type fusion splicer in a state where the cut end surfaces face each other.

In the alignment step S3, the alignment of SMF and HNLF is performed by using the alignment function of the arc discharge type welding machine. In this alignment operation, in order to prevent an increase in connection loss due to a misalignment between the core of the SMF and the core of the HNLF, core alignment is performed based on estimation of the core position by image processing, or power meter alignment is performed using a power meter.

In the welding step S4, the ends of the SMF and HNLF are melted by heating in a state of abutting against each other by arc discharge in the arc discharge type welding machine. As a result of which the butt joint portion is integrated. Hereinafter, the arc discharge power at this time is referred to as main discharge power.

In the additional discharge step S5, the arc discharge type fusion splicer further heats a certain range (the region on the HNLF20 side in the region TP shown in the configuration 0 in fig. 2) in the optical fiber longitudinal direction including the connection point between the SMF and the HNLF (corresponding to the fusion splice point 31 (32) shown in the configuration 0 in fig. 2) by the arc discharge. The length of the heated region in the longitudinal direction of the optical fiber is preferably 1 time or more the outer diameter of the cladding of HNLF. When heating is performed by this discharge, in order to confirm a change in connection loss based on a measured optical power value obtained by a power meter, a light source is optically connected to one end of an optical fiber line including the integrated SMF and HNLF, and the power meter is optically connected to the other end of the optical fiber line. Then, additional discharge is performed for a fixed time a plurality of times, and the additional discharge is repeated until the change in the connection loss at the wavelength of 1550nm before and after one additional discharge becomes 0.01dB or less (preferably 0.005dB or less). When the change in the connection loss before and after the primary additional discharge is 0.01dB or less (preferably 0.005dB or less), it is determined that the connection loss is extremely small, and the additional discharge step S5 is ended. The arc discharge power at this time is referred to as additional discharge power.

In the reinforcing step S6, the glass in the region including the fusion point between the SMF and the HNLF is covered with a recoating resin 300 (recoating) or stored in a reinforcing sleeve 400 as shown in the structure 1 in fig. 2. By this reinforcing step S6, the obtained optical fiber line can have a strength to withstand a tension of 200kpsi or more over the entire length of the optical fiber.

Fig. 7 is a table summarizing the manufacturing conditions and connection losses of the optical fiber lines of the samples of the present disclosure, and the manufacturing conditions and connection losses of the optical fiber lines of comparative examples 1 to 3, respectively. In addition, fig. 8 is a graph showing a relationship between the total additional discharge time in the additional discharge process and the connection loss of each connection point for each optical fiber line of the sample of the present disclosure, comparative example 2, and comparative example 3. In fig. 8, a curve G810 shows a discharge-loss relationship of the sample of the present disclosure, a curve G820 shows a discharge-loss relationship of comparative example 2, and a curve G830 shows a discharge-loss relationship of comparative example 3.

In the production of comparative example 1, the additional discharge step was not performed, and the process was performed until the optical fiber was connected in the fusion-splicing step. In comparative example 1, the connection loss at each welding point at a wavelength of 1550nm immediately after the welding step was 0.75dB. This connection loss is less than the ideal butt loss 2.1dB calculated from the MFD mismatch of the constant portion SP, but cannot be said to be a practically sufficiently small connection loss.

In the production of comparative example 2, the welding step and the additional discharge step were performed under the same conditions as in comparative example 1. The additional discharge power in the additional discharge step was 50% with respect to the main discharge power in the welding step. In comparative example 2 (curve G820 of fig. 8), the connection loss at each connection point at a wavelength of 1550nm immediately after the end of the fusion bonding step was 0.77dB. In the additional discharge process, the connection loss at each connection point at a wavelength of 1550nm was 0.36dB (minimum) in an additional discharge total time of 1000 seconds. It is found that the additional discharge step reduces the connection loss to less than half of the connection loss immediately after the welding step. Here, the total additional discharge time required to reduce the connection loss to one fifth or less of the ideal butting loss calculated from the MFD mismatch of the constant portion is 580 seconds. In addition, the production conditions and evaluation of each of the samples of comparative example 3 and the present disclosure will be described later.

The maximum reaching temperature in dopant diffusion by flame heating is low compared to heating by electric discharge, and therefore heating is required for one junction for about 30 minutes. In the production of comparative example 2, the dopant diffusion by the discharge heating was performed for 10 minutes or less, and the connection loss was reduced to one fifth or less of the ideal butting loss calculated from the MFD mismatch of the constant portion. In addition, in the dopant diffusion by flame heating, after the fusion-bonding process by the arc discharge fusion splicer, it is necessary to transfer the optical fiber to a dopant diffusion apparatus by flame heating. In contrast, diffusion of dopants by electric discharge heating has an advantage that the welding step and the additional discharge step can be continuously performed by the arc discharge type welding machine, and thus can contribute to improvement in productivity.

Here, if the total additional discharge time can be further shortened, the productivity can be further improved. Furthermore, when the connection loss is high, the input optical power required to obtain the same output light from the HNLF module (configuration 2 of fig. 1) needs to be increased by a corresponding amount, and the reliability of the seed light source of the HNLF module is reduced. Therefore, if the connection loss can be further reduced, the reliability of the device mounting the HNLF module can be further improved.

From the above examination, a study was made on manufacturing conditions for realizing a lower connection loss with a shorter total additional discharge time than that of the manufacture of comparative example 2. First, changes in MFD along the length direction of the optical fiber were examined for the optical fiber lines of comparative examples 1 and 2. In this study, the refractive index distribution was measured nondestructively by interferometry at each position along the length direction of the optical fiber line. Then, from the measured refractive index distribution, the field distribution of the fundamental waveguide mode is calculated by the finite element method, and the MFD is calculated.

Fig. 9 is a graph showing a change in MFD along the optical fiber length direction (the length direction of the optical fiber line) for each of the optical fiber lines of the sample of the present disclosure, comparative example 1, and comparative example 2. In addition, in fig. 9, a curve G910 represents a change in MFD of the sample of the present disclosure, a curve G920 represents a change in MFD of comparative example 1, and a curve G930 represents a change in MFD of comparative example 2. The position of 0mm in the longitudinal direction was set at a position where the optical characteristics and composition of the optical fiber are considered to be discontinuously changed (i.e., fusion point). In the longitudinal position, the negative region corresponds to the end on the HNLF side, and the positive region corresponds to the end on the SMF side. In the connection of optical fibers different in MFD from each other, it is preferable to form an MFD transition portion that realizes an MFD taper in which the MFD gradually increases from the optical fiber side where the MFD is small toward the optical fiber side where the MFD is large.

In comparative example 1 (curve G920), the MFD increased from 4.8 μm to 6.3 μm in the interval from-0.2 mm to-0.05 mm. The ideal butt loss calculated from this increased MFD is 0.8dB, which is approximately consistent with the connection loss of comparative example 1. That is, it is considered that the connection is performed with a connection loss smaller than the ideal butt loss of 2.1dB calculated from the MFD mismatch of the constant portion by realizing the MFD transition portion of the MFD cone. However, the MFD cone is not formed enough to eliminate the mismatch of the MFDs.

In comparative example 2 (curve G930), the MFD gradually increases in the section from-0.3 mm to-0.15 mm, but decreases in the section from-0.15 mm to 0mm, and an MFD transition portion in which the MFD cone is deformed is formed (a state in which the MFD does not monotonically change along the fiber length direction) can be seen. It is thus presumed that the total time for additional discharge can be shortened and the connection loss can be further reduced by forming the MFD transition portion so that the MFD cone is not deformed.

The sample of the present disclosure (curve G910) has an increased MFD even in the interval of-0.05 mm (= -50 μm) to 0mm (weld point), unlike in comparative example 1 described above. Further, the MFD of the sample of the present disclosure increases in the interval of-0.3 mm (= -300 μm) to 0 mm. As described above, according to the sample of the present disclosure, at least in the region between the first surface (welding end surface) located at the welding point and the second surface having a distance of 50 μm or more and 300 μm or less from the first surface in the end portion of the sample (HNLF), the MFD of the wavelength used monotonically increases from the second surface to the first surface. The MFD transition portion (the region corresponding to the region TP of the structure 0 in fig. 2) includes the region between the first surface and the second surface, and is set across the end of the HNLF and the end of the SMF with the welding point of the HNLF and the SMF interposed therebetween. Therefore, the total length of the MFD transition portion in the longitudinal direction of the optical fiber is 10mm or less, preferably 2mm or less. In particular, in a configuration in which the total length of the MFD transition portion is 2mm or less, the predetermined length of the portion of the MFD transition portion constituting a part of the HNLF in the optical fiber longitudinal direction is preferably 1.5mm or less.

In order to investigate the cause of forming the MFD transition portion in which the MFD cone is deformed in comparative example 2, the change in the refractive index distribution of HNLF and the change in MFD in consideration of the diffusion of the dopant due to heat were calculated. Generally, the diffusion rate of the dopant material added to the optical fiber is temperature dependent. FIG. 10 shows GeO as a representative dopant material ₂ And F, and the temperature dependence of the diffusion rate. In fig. 10, a curve G1010 shows the temperature dependence of the diffusion rate of F, and a curve G1020 shows GeO ₂ Temperature dependence of the diffusion rate. In HNLF having W-type refractive index profile as shown in FIG. 3, the core contains GeO ₂ The recess contains F. Since the diffusion speeds of these dopant materials are different from each other, the manner of change in the refractive index distribution with respect to the heating time differs depending on the heating temperature.

Let the diffusion velocity of F be v _F Adding GeO ₂ Is given as v _G . FIG. 11 shows the diffusion velocity ratio (v) _F /v _G ) A graph of the relationship between the heating time and the MFD of HNLF, assuming 20 or 30. FIG. 12 is a graph for explaining the refractive index distribution of HNLF in the heated state shown by the curves G1110 and G1120 in FIG. 11Graph of the shape change of the cloth. In fig. 11, a curve G1110 represents the diffusion velocity ratio (v) _F /v _G ) Curve G1120 represents the diffusion velocity ratio (v) =30 _F /v _G ) Relation of = 20. In addition, it is considered that the diffusion velocity ratio (v) is _F /v _G ) A diffusion velocity ratio (v) =30 ℃ corresponding to a heating temperature of 1800 DEG C _F /v _G ) And =20 corresponds to a heating temperature of 1500 ℃. Shape 1 shown in fig. 12 represents time t of both of the curve G1110 and the curve G1120 in fig. 11 ₀ The shape of the refractive index distribution of HNLF20 at time, shape 2 shown in FIG. 12, is shown in the graph G1120 from time t in FIG. 11 ₁ To time t ₃ The shape of the refractive index distribution of HNLF20 in the period (2), and shape 3 shown in FIG. 12 indicates the time t of the curve G1110 in FIG. 11 ₂ And time t of curve G1120 ₄ The shape of the refractive index profile of both HNLF 20.

As shown in FIG. 12, when HNLF originally having W-type refractive index profile (shape 1) is heated, geO contained in the core portion ₂ To the recess side, and F contained in the recess is diffused to the core side and the clad side. By such diffusion of the dopant material, the final HNLF has a monomodal refractive index profile (shape 3).

Diffusion velocity ratio (v) shown in the graph G1120 _F /v _G ) In the case of =20 (i.e., at a heating temperature of 1500 ℃), diffusion of F is dominant in the W-type refractive index profile (shape 1 of fig. 12) in the initial stage of the heating time (earlier period). This is because the diffusion velocity v of F _F BiGeO ₂ Velocity v of diffusion of _G And (4) the method is quick. After diffusion of F is completed and a monomodal refractive index distribution is obtained (shape 3 in FIG. 12), geO ₂ Becomes dominant. In the diffusion velocity ratio (v) _F /v _G ) In the case of =20, MFD becomes maximum 7.5 μm (the refractive index distribution is deformed as in shape 2 of fig. 12) when the refractive index distribution is changed from the W-type refractive index distribution to the unimodal refractive index distribution, that is, when the heating time exceeds 15 minutes, and then there is a time region in which MFD decreases with the elapse of the heating time. This is because the MFD transition portion where the MFD cone is deformed is formed.

One of the methods for avoiding the MFD cone deformation is to increase the F concentration in the depressed portion, thereby increasing the maximum value of MFD by extending the time at which F diffusion ends. However, the dispersion value, which is one of important optical characteristics of HNLF, is a value that varies according to the refractive index distribution of the optical fiber. Due to the change of F concentration and delta in the recess _dep Is relevant, so the F concentration is limited. That is, the diffusion of the limited F contained in the depressed portion must be controlled to avoid the deformation of the MFD cone.

On the other hand, the following findings were obtained: at diffusion velocity ratio (v) _F /v _G ) In the case of =30 (that is, at a heating temperature of 1800 ℃), the time for transition from the W-type refractive index distribution to the unimodal refractive index distribution can be shortened, and the MFD can monotonically increase with the elapse of the heating time. That is, it is considered that the deformation of the MFD cone can be avoided by raising the heating temperature without increasing the F concentration in the depressed portion. Obviously, the diffusion velocity ratio (v) capable of thus avoiding deformation of the MFD cone _F /v _G ) The critical value of (d) is between 20 and 30, and in the samples of the present disclosure, the diffusion velocity ratio (v) will be _F /v _G ) Is set to 22 or more and 40 or less.

Based on this finding, welding was performed under the discharge conditions performed in the sample production of the present disclosure (see fig. 7 and 8). As shown in fig. 7, the welding step performed in the sample production of the present disclosure was performed under the same conditions as in comparative examples 1 and 2. In addition, the additional discharge power in the additional discharge process performed in the manufacture of the sample of the present disclosure was 65% with respect to the main discharge power in the fusion bonding process performed in the manufacture of comparative example 1 (a power 15% higher than the additional discharge power in the additional discharge process performed in the manufacture of comparative example 2).

As a result, as shown in fig. 8, in the sample (curve G810) of the present disclosure, the connection loss at each connection point at a wavelength of 1550nm immediately after the end of the welding process was 0.73dB, which was substantially equal to that of comparative example 2. In the additional discharge process carried out in the sample fabrication of the present disclosure, the connection loss at each connection point of 1550nm wavelength was 0.24dB (minimum) with an additional discharge total time of 500 seconds. In the sample of the present disclosure, the connection loss can be further reduced by 0.12dB with half the additional total discharge time, as compared to comparative example 2.

In addition, in the case of the sample of the present disclosure, the total additional discharge time required to reduce the connection loss to less than one fifth of the ideal butt loss expressed in dB, which is calculated from the MFD mismatch of the respective constant portions of SMF and HNLF, is 200 seconds. The F area concentration in the valleys of HNLF added to the samples of the present disclosure was 1.6 x 10 ⁶ ppm·μm ² . The range of F area concentration is preferably 0.4X 10 ⁶ ppm·μm ² Above and 3.2X 10 ⁶ ppm·μm ² Hereinafter, more preferably 0.8 × 10 ⁶ ppm·μm ² Above and 2.4X 10 ⁶ ppm·μm ² The following.

Similarly to the case of comparative example 1, the MFD change in the fiber longitudinal direction at the MFD transition portion (the region formed in the region TP in configuration 0 of fig. 2) of the optical fiber line of the sample of the present disclosure was examined. The results are shown in fig. 9. The sample of the present disclosure is formed with an MFD cone that monotonically increases from the position of-0.3 mm to the position of 0mm, as compared with comparative example 1. Therefore, at least reheating in the additional discharge step performed in the sample production of the present disclosure needs to be performed for 50 seconds or more at a discharge power at which the F (fluorine) contained in the HNLF is diffused without melting the HNLF cladding layer.

On the other hand, in the production of comparative example 3, as shown in fig. 7, the additional discharge power was set higher than that of the sample of the present disclosure, and the additional discharge power was 75% with respect to the main discharge power of comparative example 1. The welding step performed in the production of comparative example 3 was performed under the same production conditions as in any of comparative examples 1 and 2 and the samples of the present disclosure. As a result, as shown in fig. 8, in comparative example 3 (curve G830), the connection loss became 0.56dB (minimum) at the time of the additional discharge total time of 200 seconds, and thereafter, the connection loss increased. This is considered to be because if the additional discharge power is set too high, the optical fiber is heated to a temperature at which the optical fiber is deformed in the heated region during the additional discharge, and as a result, it is difficult to reduce the connection loss. That is, in the case of comparative example 3, the connection loss cannot be reduced to one fifth or less of the ideal butting loss calculated from the MFD mismatch of the constant portion.

The high fracture strength of the weld may be an important factor in ensuring long-term reliability of the module. Fig. 13 is a graph showing the results of evaluating the mechanical strength characteristics of the optical fiber line of each of the sample of the present disclosure and the comparative example 2 (the tensile strength test results of the fusion-spliced portion of the sample of the present disclosure and the comparative example 2) as a weibull distribution curve. In fig. 13, the symbol "□" represents the evaluation result relating to the sample of the present disclosure, and the symbol "good" represents the evaluation result relating to comparative example 2. The straight line L1 is an approximate straight line of the evaluation results of the sample of the present disclosure obtained by the least square method, and the straight line L2 is an approximate straight line of the evaluation results of comparative example 2. The inclination of the straight line L1 and the straight line L2 indicates the deviation of the evaluation result, and a larger inclination means a smaller deviation of the evaluation result.

As can be seen from fig. 13, in the sample of the present disclosure, the 50% fracture strength (fracture strength when the cumulative fracture probability F = 0.5) is 7.5N (0.77 kgf). On the other hand, in the case of comparative example 2, the 50% breaking strength was 4.5N (0.46 kgf). Therefore, the 50% breaking strength in the tensile strength test is preferably 4.5N or more. It is considered that the sample production of the present disclosure can shorten the total additional discharge time compared to the case of producing comparative example 2, and thus can achieve high fracture strength. That is, the optical fiber line of the sample of the present disclosure can achieve both low connection loss and high breaking strength.

Description of the reference numerals

10. 11, 12: SMF (single mode fiber), 20: HNLF (high nonlinear fiber), 31, 32: welding point (welding end face), 50: metal case, 101, 102: core, 202: recess, 102, 203: cladding, 103, 204: coating resin layer, 300: recoating resin, 400: reinforcing sleeve, TP: MFD taper to region (consistent with MFD transition), SP: a constant part.

Claims

1. An optical fiber circuit comprising:

a high nonlinearity optical fiber, comprising: a core, a depressed portion surrounding the core and having a refractive index lower than that of the core, and a clad surrounding the depressed portion and having a refractive index lower than that of the core and higher than that of the depressed portion,

a single mode optical fiber fusion spliced to the high nonlinearity fiber and comprising: a core, and a cladding surrounding the core and having a refractive index lower than that of the core, an

An MFD converter which is configured by end portions of the high nonlinear optical fiber and the single mode optical fiber butted at a fusion point of the high nonlinear optical fiber and the single mode optical fiber, and is defined as a section in which a mode field diameter of a wavelength used changes so that a difference between a maximum value and a minimum value is 0.3 [ mu ] m or more over a distance of 100 [ mu ] m in an optical fiber length direction,

a connection loss of the high nonlinear optical fiber and the single mode optical fiber at a wavelength of 1550nm is less than one fifth of an ideal butt loss calculated based on a mismatch between mode field diameters at a wavelength of 1550nm of a constant portion other than the MFD transition portion of both the high nonlinear optical fiber and the single mode optical fiber,

the MFD transition portion has a total length defined along the longitudinal direction of the optical fiber of 10mm or less,

in a region of the end portion of the high nonlinear optical fiber between a first surface located at the fusion-splice point and a second surface located at a distance of 50 μm or more and 300 μm or less from the first surface, a mode field diameter of the use wavelength monotonically increases from the second surface toward the first surface,

a diffusion velocity v of fluorine in the region between the first face and the second face at a distance of 50 μm from the first face in the end portion of the high nonlinearity fiber _F And GeO ₂ Of (d) diffusion velocity v _G Ratio of (v) _F /v _G ) Is 22 or more and 40 or less.

2. The fiber optic line of claim 1,

the highly nonlinear optical fiber is composed of silica glass,

in the constant portion of the high nonlinear optical fiber, the core does not contain a refractive index lowering agent as a dopant for lowering a refractive index of glass at a wavelength of 1550nm and contains a refractive index raising agent as a dopant for raising the refractive index of glass at a wavelength of 1550nm, the depressed portion does not contain the refractive index raising agent and contains the refractive index lowering agent, and the cladding does not contain the refractive index lowering agent.

3. The fiber optic line of claim 2,

in the constant portion of the high nonlinear optical fiber, the core contains GeO as the refractive index increasing agent ₂ The depressed portion contains the refractive index lowering agent in an area concentration of 0.4X 10 ⁶ ppm·μm ² Above and 3.2X 10 ⁶ ppm·μm ² The following fluorine.

4. The fiber optic line of any one of claims 1-3,

in the end portion of the high nonlinearity fiber, an outer diameter of the cladding of the high nonlinearity fiber monotonically decreases or increases from the constant portion of the high nonlinearity fiber to the constant portion of the single mode fiber.

5. The fiber optic line of any one of claims 1-3,

the total length of the MFD transition portion is 2mm or less,

the MFD transition portion has a portion constituting a part of the high nonlinear optical fiber, and a predetermined length in the longitudinal direction of the optical fiber is 1.5mm or less.

6. The fiber optic line of any one of claims 1-3,

the single mode optical fiber is an optical fiber conforming to the ITU-T g.657.A1 standard.

7. The fiber optic line of any one of claims 1-3,

in the constant portion of the high nonlinear optical fiber, an outer diameter of the core is 2.5 μm or more and 4.0 μm or less, an outer diameter ratio of the depressed portion to the core is 1.8 or more and 3.2 or less, a relative refractive index difference between a maximum refractive index of the core and an average refractive index of the cladding is 1.3% or more and 2.0% or less, a relative refractive index difference between a minimum refractive index of the depressed portion and the average refractive index of the cladding is-1.0% or more and-0.5% or less,

the constant portion of the high nonlinear optical fiber has:

10 μm at 1550nm wavelength ² Above and 30 μm ² The effective area of the following is,

the mode field diameter at a wavelength of 1550nm of 3.5 μm or more and 5.0 μm or less,

at 1550nm wavelength-3.0 ps km ^-1 ·nm ^-1 Above and 1.5ps km ^-1 ·nm ^-1 The following chromatic dispersion, and

at 1550nm wavelength of 7W ^-1 ·km ^-1 Above and 20W ^-1 ·km ^-1 The following nonlinear coefficients.

8. The fiber optic line of any one of claims 1-3,

a butt joint region constituted by the end portion of the high nonlinear optical fiber and the end portion of the single mode optical fiber and including the fusion-splice point is protected by being covered with a resin or being housed in a reinforcing sleeve,

the butt-joint region protected by the resin or the reinforcing sleeve has a strength of 200kpsi or more.

9. The fiber optic line of any one of claims 1-3,

at least one of the high nonlinear optical fiber and the single mode optical fiber has a covering resin layer surrounding the cladding and containing a colorant,

in the structure in which both the high nonlinear optical fiber and the single mode optical fiber have the covering resin layers, the covering resin layers of the high nonlinear optical fiber and the single mode optical fiber contain colorants different in color from each other, respectively.

10. The fiber optic line of any one of claims 1-3,

the 50% breaking strength of the optical fiber line in a tensile strength test is 4.5N or more.

11. A module, comprising:

the fiber optic line of any one of claims 1-10; and

and a metal housing which houses the optical fiber circuit.

12. A method of manufacturing an optical fiber circuit, comprising:

an arrangement step of arranging end portions of single-mode optical fibers facing end portions of highly nonlinear optical fibers in a butted state, the highly nonlinear optical fibers including: a core, a depressed portion surrounding the core and having a refractive index lower than that of the core, and a clad surrounding the depressed portion and having a refractive index lower than that of the core and higher than that of the depressed portion,

a fusion-splicing step of fusing the end portions of the highly nonlinear optical fiber and the single-mode optical fiber by heating the end portions of the highly nonlinear optical fiber and the single-mode optical fiber that are butted to each other after the arranging step; and

an additional discharging step of reheating by discharging a region located on the highly nonlinear optical fiber side with respect to a fusion-splicing point of the highly nonlinear optical fiber and the single-mode optical fiber after the fusion-splicing step, in order to manufacture the optical fiber line according to any one of claims 1 to 10.

13. The optical fiber circuit manufacturing method according to claim 12,

in the additional discharge step, the region located on the high nonlinear optical fiber side with respect to the fusion-splicing point is continuously reheated for 50 seconds or more at a discharge power for diffusing fluorine contained in the high nonlinear optical fiber without melting a cladding of the high nonlinear optical fiber.

14. The optical fiber circuit manufacturing method according to claim 12 or 13,

in the additional discharge step, the end portion of the high-nonlinearity fiber and the end portion of the single-mode fiber are reheated by discharge while a heating portion is moved relative to the end portion of the high-nonlinearity fiber and the end portion of the single-mode fiber, which are fusion-spliced, respectively.

15. The optical fiber circuit manufacturing method according to claim 12 or 13,

in the additional discharge step, reheating by discharge is performed in a range in which a length in a longitudinal direction of the optical fiber is 1 time or more of an outer diameter of the clad.